Selinexor in Uveal Melanoma: A Hopeful Step or Just Another Hurdle? Open Access

Metastatic uveal melanoma (UM) is a rare cancer with an incidence of 5.1 cases per million per year in the United States, although it is the most frequent intraocular malignancy in adults.^[1] Despite advances in oncology treatments over the last decade, they have been relatively scarce in metastatic UM. Therefore, the prognosis remains poor with a median overall survival (OS) below 1 year for patients with stage IV disease.^[2] In recent years, new systemic therapies have been approved in the first- and second-line treatment setting.

Two recent studies have placed nivolumab plus ipilimumab as first-line treatment in patients with metastatic UM. The GEM-1402 trial, an open-label, single-arm, phase II study by the Spanish Multidisciplinary Melanoma Group, studied the anti-PD1–anti-CTLA4 combination in 52 treatment-naïve patients with stage IV UM.^[3] This study met its primary endpoint of 12-month OS (51.9%; 95% CI, 38.3–65.5%), with median OS and progression-free survival (PFS) of 12.7 and 3 months, respectively. Still, the objective response rates (ORRs) were very low (11.5%), with one complete response and five partial responses.^[4] Similarly, another US single-arm phase II study explored the combination of nivolumab plus ipilimumab in 35 patients with metastatic UM, including systemic treatment-naïve (57% of patients) and pretreated patients.^[5] The 12-month OS was 56% (95% CI, 38–71%), consistent with results reported in the Spanish GEM-1402 trial,^[3] although ORR remained low (18%, with only one patient experiencing complete response).^[5]

Recently, a phase III trial has added tebentafusp to the stage IV UM treatment picture.^[6] Tebentafusp is a bispecific protein with affinity to HLA-A*02:01–restricted T-cell receptor, specific for glycoprotein 100, and fused to an anti-CD3 effect, attracting T cells to target glycoprotein 100–positive cells. This study, although comparing experimental treatment to single-agent pembrolizumab, ipilimumab, or dacarbazine, reported a 12-month OS of 73% (95% CI, 66–79%).^[6] Tebentafusp is not an option for every patient, as approximately 45% of people are HLA-A*02:01 positive. Double-agent immunotherapy and tebentafusp have modified metastatic UM first-line treatment; however, efficacy results are modest, and new therapies are needed.

Selinexor (KPT-330) is an orally available irreversible inhibitor of exportin 1 (XPO1).^[7] XPO1 is responsible for the transportation of nuclear proteins such as p53, RB1, and p27 to the cytoplasm.^[8] Overexpression of XPO1 has been associated with development of various solid and hematologic therapies. In addition, XPO1 plays a role in the development of drug resistance to neoplasm therapies.^[9] Selinexor selectively blocks XPO1, avoiding the export of tumor suppressor proteins to the cytoplasm, thus accumulating them in the nucleus and leading to cell apoptosis (Fig. 1).^[10]

A phase 1B, single-center, open-label, multi-arm trial (ClinicalTrial.gov ID: NCT02419495) investigated selinexor in association with other treatments (either immunotherapy or chemotherapy) in different solid tumors, using a standard 3 + 3 design and a “basket-type” expansion.^[11] Multiple publications on this trial have reported a good safety profile across cancer types.^[12–14] Alhalabi et a^[12] have reported dose-limiting toxicities in 2of 14 patients and an ORR of 19% in patients with renal cell carcinoma treated with selinexor in association with immune-checkpoint inhibitors (ICIs) (nivolumab alone or in association with ipilimumab). Gouda et al^[13] reported that 25 patients with advanced (nonuveal) melanoma who were receiving selinexor with pembrolizumab showed a high incidence of adverse events (96%), although most treatment-related toxicities were manageable. The ORR in 10 patients with treatment-naïve evaluable disease was 70%, significantly higher than that of the 14 patients with prior ICI treatment (7%).^[13] In another recent publication on this trial, Gouda et al^[14] focused on advanced and metastatic UM treated with selinexor and ICIs. Patients with advanced or metastatic UM were enrolled to receive selinexor with either pembrolizumab or nivolumab plus ipilimumab. The primary endpoint of this phase 1B study was to evaluate the safety profile. Ten patients with metastatic UM were recruited: nine patients received selinexor plus pembrolizumab and one received selinexor plus nivolumab and ipilimumab. Four patients had been previously treated with systemic treatment; among them, three received ICI therapy in a previous treatment line setting. A dose reduction was required in 60% of cases owing to toxicity. Although the study demonstrated a manageable safety profile, its limited sample size and absence of a control arm warrant caution before drawing strong clinical conclusions. On the other hand, when comparing the data from this study with the report on patients affected by advanced nonuveal melanoma, a similar proportion of patients required dose reduction owing to toxicity (56%).^[13] The rate of treatment discontinuation due to toxicity was different between the two trials, with 33% for patients with UM and 17% for patients treated for melanoma. The most common adverse events were neutropenia (70%), thrombocytopenia (60%), leukopenia (50%), and anemia (50%); the most common grade 3 or more toxicity was a decrease in neutrophil count (50%), with other toxicities inferior to 10% of cases; no other specific drug-related toxicities were reported.^[13,14] Gouda et al^[13] reported stable disease (SD) as the best response in 89% of evaluable patients with UM (8/9), with one patient with progression; the range of SD was very wide (1–30 months) with an average of 5 months.^[13] However, no ORRs (partial or complete remission) were observed, raising concerns about the clinical relevance of the findings. SD is a measure that may not always translate to long-term survival benefits, particularly in a disease where median PFS is typically short, with 6 months median PFS in this specific case. This study reports a median OS of 17 months and a 12-month OS of 55.6%.

Comparing these results with previously reported outcomes of ICI treatments, we consider the 1-year OS rate with immunotherapy to be 52% in the first-line setting and 56% in any treatment line setting.^[4,5] This is relevant because it demonstrates that there is no substantial difference in immunotherapy efficacy, based on the treatment line in which ICIs are administered. In Gouda et al’s studies,^[13,14] some patients received immunotherapy earlier than others, and this did not appear to influence the outcome. Indeed, when evaluating patient outcomes (stable or progressive disease) based on the number of prior treatment lines, SD was achieved not only in the first-line setting but also in the setting of two patients who received the treatment after three prior lines of therapy. This combination is certainly promising as a potential therapeutic alternative; however, it remains unclear whether selinexor provides any additional survival benefit when compared to immunotherapy alone. Furthermore, only one patient received the combination of nivolumab and ipilimumab alongside selinexor. Although the treatment with nivolumab plus ipilimumab and the treatment with pembrolizumab were part of two different arms of the study, the studies do not specify the reason for this discrepancy between the two treatments. This raises concerns regarding potential prior interactions or selection biases that may have influenced the treatment allocation. Further clarification is needed to understand the underlying reasons for this difference. When examining data for tebentafusp, the 1-year OS rate was 73%, with a median OS of 21.6 months and a PFS of approximately 3.4 months.^[6,15] These findings underscore the key role of these two endpoints, PFS and OS, when analyzing clinical trials in patients affected by solid tumors with poor prognosis.

As reported in the main article, the study has some limitations such as small sample size; single center nonrandomized recruitment, which may not capture the diversity of the population; and the lack of correlative biomarkers in this cohort. However, considering the reported synergy of selinexor in combination with chemotherapy in other malignancies,^[16] trials evaluating selinexor in combination with tebentafusp or chemotherapy might offer superior efficacy when compared to ICIs alone. This phase 1B trial provides preliminary safety data on the combination of SEL and ICI therapy in UM; however, the lack of objective responses, the toxicity burden, and the small sample size limit immediate clinical applicability. As reported in a study by Vergote et al,^[16] a specific cohort, such as patients with TP53wt endometrial cancer, may benefit more from selinexor than the overall endometrial population. This could explain the significant variability in treatment benefits observed in this study (SD range, 1–30 months). Therefore, increasing the number of patients is essential to identify potential subpopulations that may benefit more from selinexor treatment. Furthermore, the use of liquid biopsy, namely circulating tumor DNA (ctDNA), might be a useful treatment biomarker of efficacy and might identify patients benefitting most from this combination, as ctDNA clearance was strongly associated with OS in UM, even in patients with radiographic progression, in a phase II study with tebentafusp.^[17] These results resemble those reported in other disease states and might support the use of liquid biopsy in the evaluation criteria for response, as recently proposed (LB-RECIST).^[18] Finally, larger randomized trials with biomarker-driven patient selection are needed to determine whether this combination has a role in the treatment paradigm of UM.